The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute specific Python scripts, including manage_markdown_site.py and manage_markdown_work.py, to build, validate, and manage the delivery of SEO audit documents. The location of these scripts is resolved dynamically at runtime through a configuration file named .seo-suite-runtime.json.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and audit external content, such as web page captures and Research Packs. This audit surface is susceptible to instructions embedded within the audited content that could attempt to manipulate the agent's findings or recommendations.
Ingestion points: The agent is instructed to read research-pack.json and pinned target captures (HTML/source) as part of the procedure defined in SKILL.md.
Boundary markers: The skill requires strict SHA-256 hash verification for all inputs and metadata to ensure immutability, although it does not define specific prompt-level delimiters to isolate the content being audited from the agent's instructions.
Capability inventory: The agent has the capability to write substantive Markdown reports (aeo-findings.md, answer-map.md), generate JSON optimization briefs, and execute local suite-provided Python scripts.
Sanitization: The skill implements a robust audit-protocol.md and a 'Myth-resistance gate' designed to identify and reject unsupported engine-mandatory rules, such as fixed passage lengths or universal AEO scores, which serves as a logical layer of sanitization against common SEO manipulation tactics.

seo-aeo